Jump to content


Skill based match making


  • Please log in to reply
73 replies to this topic

leeuniverse #61 Posted Jul 30 2019 - 04:59

    Major

  • Players
  • 39059 battles
  • 8,063
  • [LAMP] LAMP
  • Member since:
    01-30-2013

View PostMacDaddyMatty, on Jul 29 2019 - 06:29, said:

You ignored the question.

I'm Blue in the Emil 2, Amx 50 120, T28 JHTC.

I'm Red in the Bulldog, AMX ELC Bis.

Work that into your MM.

 

You clearly didn't read what I wrote.

Teams are balanced by Player Rating, which is your OVERALL skill playing the game playing all tanks.

So, just because you are one way or another with a particular tank it doesn't in fact matter, because everyone else will also be in the same situation.

 

Also, if you further know what I write on the subject, you know that I've said over and over that SBMM will only provide a "minor" improvement in game balance, but it's also a necessary first step.

LOT's of other things as I've already stated are also needed to be done.

So, you're not talking to someone who thinks SBMM is a "cure-all".  It's not going to be.  At most it would likely just be at most a 5% improvement in creating balanced games, and that's likely the high end.

Again, much more is needing to be done.

 

So again, I didn't "ignore" anything in your post, but YOU certainly did.



Hellsfog #62 Posted Jul 30 2019 - 05:13

    Major

  • Players
  • 38278 battles
  • 6,804
  • Member since:
    06-22-2011

View PostNeatoMan, on Jul 25 2019 - 08:45, said:

 

snip

 

 The test server also has players playing tanks and tiers they are unfamiliar with, and doing things they normally wouldn't try on the live server..  People take far more risks, and are probably far more aggressive on the test server.

 

 This I can agree with for sure, as well as buffing and nerfing UP/OP tanks properly.   They have never gotten this part right.

 

 This ^  Besides, can anyone get data from the test server or are we relying on lee's feelings again?



MacDaddyMatty #63 Posted Jul 30 2019 - 12:11

    Major

  • -Players-
  • 31936 battles
  • 6,925
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View Postleeuniverse, on Jul 29 2019 - 22:59, said:

 


 

Teams are balanced by Player Rating, which is your OVERALL skill playing the game playing all tanks.


 

THIS IS FOLLY AND MEANS NOTHING.

If someone pads and grinds, they are going to have inflated stats.

Many only play SHs / CW, and only occasional pubs - My stats in the WZ-111 5A are terribad in pubs because I only run it in CW where I do quite well. This is not factored in my player rating!

Overall is not particularly useful in the nuanced nature of the game.

Player skill in THAT tank MM is looking at are not being factored in your metrics.


 

What about platoons?

Mixed platoons?


 

At least address and justify your reasoning for SOME of that.

PLEASE!


 

If it was a good idea we would see it.


 

SBMM is a waste when the whole team of various colors all go valley on Lakeville.



leeuniverse #64 Posted Jul 30 2019 - 14:08

    Major

  • Players
  • 39059 battles
  • 8,063
  • [LAMP] LAMP
  • Member since:
    01-30-2013

View PostMacDaddyMatty, on Jul 30 2019 - 04:11, said:

THIS IS FOLLY AND MEANS NOTHING.

If someone pads and grinds, they are going to have inflated stats.

Many only play SHs / CW, and only occasional pubs - My stats in the WZ-111 5A are terribad in pubs because I only run it in CW where I do quite well. This is not factored in my player rating!

Overall is not particularly useful in the nuanced nature of the game.

Player skill in THAT tank MM is looking at are not being factored in your metrics.

 

What about platoons?

Mixed platoons?

 

At least address and justify your reasoning for SOME of that.

PLEASE!

 

If it was a good idea we would see it.

 

SBMM is a waste when the whole team of various colors all go valley on Lakeville.

 

1. You're confusing WN8 with Player Rating.  They are not the same at all, even though they do have general Trends of similarity like Unicorns being in the 8-10,000 Player Rating range, Blues being 7-9,000, etc. etc.

2. SBMM by "Winrate" is what's Unreliable, not by Player Rating.

3. Again, I don't care if you're bad in some tanks in Random, you're clearly going to be good in some, so Player Rating takes into account your overall skill in multiple respects from spotting, to damage, etc. etc.

4. LOL, I've already justified everything.  I've already said SBMM will only be a MINOR improvement in game balance.

5. Again, I'm well aware SBMM won't solve many problems, including Lemming Trains in certain instances, including positive iterations of it which creates Steamrolls.

6. Again, SBMM adds NOTHING to "waist" in relation to the MM etc.  It simply balances each team by what's in the que as equally as possible with a similar # of similar skilled players at ALL levels of skill.

 

If it doesn't "hurt" anything, you should ask yourself why you're so "offended" by it?

You also don't address all the other things I've stated also need to be done to see some real improvement in balancing games.

How are YOU going to be "hurt" by the game being made better for more people?



NeatoMan #65 Posted Jul 30 2019 - 14:52

    Major

  • Players
  • 28180 battles
  • 20,546
  • Member since:
    06-28-2011

View Postleeuniverse, on Jul 30 2019 - 08:08, said:

1. You're confusing WN8 with Player Rating.  They are not the same at all, even though they do have general Trends of similarity like Unicorns being in the 8-10,000 Player Rating range, Blues being 7-9,000, etc. etc.

The various metrics pretty much mirror each other, so not much difference there



MacDaddyMatty #66 Posted Jul 30 2019 - 14:54

    Major

  • -Players-
  • 31936 battles
  • 6,925
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

1. NO, I am not. I am looking at your proposal of player rating which take all tanks / all tiers and combines them resulting in a score. So the guy w/every other match is in his tier 2, six perk crew AWESOME PR tank with bond equipment, and every other match is grinding his AWFUL PR tank; looks like he has an average PR to your MM.


 

4. Then the Juice is not worth the Squeeze IMHO.


 

Not offended by anything other than the ignored fact that IF SBMM could work (which I am unconvinced it would ever accomplish anything other than long wait times as ppl, drop out of cue and rejoin due to "wait frustration"), then SBMM proponents are advocating that everyone is funneled towards a 50% winrate, punctuated by:

- unicums being the last tanks on each team;

- in battles that go on the entire length of the match;

- locking up the tanks we are trying to grind;

- wasting our multipliers / boosters while we wait for the timer to run out.


 

We have VERY different definitions of fun.



leeuniverse #67 Posted Jul 30 2019 - 21:59

    Major

  • Players
  • 39059 battles
  • 8,063
  • [LAMP] LAMP
  • Member since:
    01-30-2013

View PostMacDaddyMatty, on Jul 30 2019 - 06:54, said:

1. NO, I am not. I am looking at your proposal of player rating which take all tanks / all tiers and combines them resulting in a score. So the guy w/every other match is in his tier 2, six perk crew AWESOME PR tank with bond equipment, and every other match is grinding his AWFUL PR tank; looks like he has an average PR to your MM.

 

4. Then the Juice is not worth the Squeeze IMHO.

Not offended by anything other than the ignored fact that IF SBMM could work (which I am unconvinced it would ever accomplish anything other than long wait times as ppl, drop out of cue and rejoin due to "wait frustration"), then SBMM proponents are advocating that everyone is funneled towards a 50% winrate, punctuated by:

- unicums being the last tanks on each team;

- in battles that go on the entire length of the match;

- locking up the tanks we are trying to grind;

- wasting our multipliers / boosters while we wait for the timer to run out.


 

We have VERY different definitions of fun.

 

1. Everyone else is in the same boat, we ALL have tanks we are better and worse in, so it doesn't matter.

2. There is no "squeeze"...  But there's some juice, so it's worth it.

Did you even see my post that if all of my suggestions are followed, balanced games will occur likely anywhere between 36-60% more often?

Then say imbalanced games occur 40% of the time (or say it's 60%), 36-60% of that, i.e. imbalanced games removed, is a significant improvement in game quality of life for players.

3. This has been said over and over again, yet somehow you're still not getting it.

There is ZERO ADDED WAIT TIME TO THE QUE...  All the SBMM that we are talking about does is simply the que picking out the players it currently does, and instead of placing one team with more better players, it places them in an EQUAL # of similar skilled on each team.  And btw, those who have done studies by WN8 Totals and thinking SBMM wouldn't make a difference because WN8 totals aren't hugely different, aren't considering that a team with 6 Blues can have a similar WN8 total as a team of 8 Greens (for example), yet a 5 Blue team will almost always win, so those who did those studies were using flawed methodology.  So, what I'm doing is placing an equal # of Blues and Greens on each team, and other levels also.

 

None of what you just stated is true because it's based on a false premise.  My premise is the above.

 

So, my definition of fun, is tanks performing well, aka move, shoot, etc. non-frustratingly, matches being long-lasting, which means things that contribute to creating short games, steamrolls etc. are greatly reduced (which all my improvements will do), so a persons game isn't WHAM!!!! and you're games over, so people don't feel like they've wasted their time, and thus end up QUITING the game due to this primary frustration, whatever the symptom be it Arty, Steamrolls, Gold Rounds, High Damage, Power Spread between tiers of the same class, etc. etc.



MacDaddyMatty #68 Posted Jul 31 2019 - 14:37

    Major

  • -Players-
  • 31936 battles
  • 6,925
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

You have left me thoroughly unconvinced!

This is still a terribad idea that AW tried, and was another nail in the coffin of that game.


 

Squeeze is the coding and work to implement.

Squeeze is refining the algorithm.

Squeeze is the costs involved both in money, manpower, and time.

Juice is the fruits of the labor.


 

Juice. Not. Worth. The. Squeeze.


 

I'm a game developer. It is clear that you don't get the processes involved.

Nor the secondary or tertiary effects of this proposal.

Your metrics are based on conjecture, and therefore meaningless.

Any change adding load to MM affects time.


 

Your premise is akin to saying that the game of Poker is unfair because someone may end up with a hand that has 3 aces, while someone else ends up with a hand that has low card.

Um... YEAH that's the game.

It's not the cards, it's what you do with them.

Play your hand well is the game.

The fix is NOT give everyone 1 Ace, 1 Jack, 1 8, etc... that game would suck.

The fix is get rid of the ability for people to see the skill of the other team.


 

Fortunately WG does and has said SBMM is muey no Bueno (ochen ne horosho).


 

Available to you that may meet your needs are tournaments, SHs, Advances, etc.

Though not SBMM, you are free to dictate the composition of your teams.



Machisman #69 Posted Jul 31 2019 - 15:04

    Sergeant

  • Players
  • 36958 battles
  • 111
  • Member since:
    10-27-2011

If MM is completely random, then make it full random. Why even have platoon? How many top tier clans we do we have in NA? may be max 10? You know how platoon get matched up right? Time and Time again weak Platoon get matched up against Top clan platoons and it completely changes the outcome of the game.

If we are talking about random, then there should not be any platoon to alter the Randomness of MM.

All we want is, once the system identifies 30 players, let is sort based on PR and assign it to each team.

 

Now a days MM is broken or it may just be that People dont play this game the way it should be. If enemy team has an EBR105, thats it, everyone wants to camp. None of the other tanks can take map control because they get spotted before going anywhere. 

For the past 4 months, my damage ratio has increased, however my Winrate is down the drain. If you want go and check websites that shows my stat history.

This game is not enjoyable anymore, however i keep playing this. Two reasons - i do not want to waste my premium time and secondly the defeats make me keep trying and not give up. 

 

However since i am playing with other 14 humans, i cant control their actions and may as well give up.

 

The game is not competitive. If only atleast half of my team knew how to play and where to go on any give map, Read the enemy team layout and decide on how to play your heavy or mediums. So many factors goes into this game. I guess NA server skill level is going down. Some of my ingame friends, who have more than 8000 PR has already quit and i did not see them online for almost two months.

 

Who ever is supporting this MM, keep doing it and wait for all the average ones to abandon this game and you might as well enjoy playing against BOTS. 

 

 

 



Jer1413 #70 Posted Jul 31 2019 - 16:01

    Captain

  • Players
  • 47621 battles
  • 1,528
  • [RR13] RR13
  • Member since:
    02-24-2013

View PostMachisman, on Jul 31 2019 - 10:04, said:

 

This game is not enjoyable anymore, however i keep playing this. Two reasons - i do not want to waste my premium time and secondly the defeats make me keep trying and not give up. 

 

 

 

 

 

 

 

So you don't find it more of a waste to play a game you don't enjoy?

 

 



Machisman #71 Posted Jul 31 2019 - 18:44

    Sergeant

  • Players
  • 36958 battles
  • 111
  • Member since:
    10-27-2011

View PostJer1413, on Jul 31 2019 - 09:01, said:

 

 

So you don't find it more of a waste to play a game you don't enjoy?

 

 

Not a waste until my premium time runs out, which is not far off. Another 15 days more. Decided not to spend even a penny anymore. My WOT times have already reduced significantly and i have started some new games. No one cares if i play or not anyways and the same goes to others.

But i will be on the list who quit due to absurd MM. Yes MM is the primary reason for moving away from this game. Wish there was a survey with this question during uninstalling the game. 



Jer1413 #72 Posted Jul 31 2019 - 21:42

    Captain

  • Players
  • 47621 battles
  • 1,528
  • [RR13] RR13
  • Member since:
    02-24-2013

View PostMachisman, on Jul 31 2019 - 13:44, said:

Not a waste until my premium time runs out, which is not far off. Another 15 days more. Decided not to spend even a penny anymore. My WOT times have already reduced significantly and i have started some new games. No one cares if i play or not anyways and the same goes to others.

But i will be on the list who quit due to absurd MM. Yes MM is the primary reason for moving away from this game. Wish there was a survey with this question during uninstalling the game. 

 

I just don't understand why you would continue to play a game you don't enjoy. What does premium time have to do with anything if you don't plan on continuing?

 



low_bidder #73 Posted Jul 31 2019 - 22:19

    Sergeant

  • -Players-
  • 24063 battles
  • 245
  • Member since:
    06-23-2015
WOT should add EBMM (equipment base match making) into the mix along with SBMM.  Not much of a chance for rookies trying to complete against tanks that have been fully equipped with enhanced equipment upgrades.

leeuniverse #74 Posted Aug 01 2019 - 01:02

    Major

  • Players
  • 39059 battles
  • 8,063
  • [LAMP] LAMP
  • Member since:
    01-30-2013

View PostMacDaddyMatty, on Jul 31 2019 - 06:37, said:

You have left me thoroughly unconvinced!

This is still a terribad idea that AW tried, and was another nail in the coffin of that game.

 

You keep stating falsehoods that have already been long debunked....

 

AW didn't do SBMM, they did WBMM, aka MM by WinRate not "skill".

Winrate MM is the worst way to do MM because there are specific classes that have worst winrates such as Arty or Lights, and if you play those classes more your Winrates are going to be lower, even if you're actually a good player in those classes.

Winrates can also be affected in several other ways that don't have to do with skill, for example, someone who likes to go "sit" on the Enemy Base well before it's known who's going to win, thus artificially inflating their Winrate.

 

The other problems with AW is that their PVP mode still had MOST of the biggest problems that people hated and left WOT for.

Most of which relate to the "Quick Death" whatever the "symptom" be, low HP values, poor armor values/structuring, power spread between tiers of the same class, etc. etc.

Then you have the great 2.0 whatever update and Obsidians firing which totally sealed the game's failure, which is when I left it, where they massively nerfed many aspects of the game, making things work worse, etc.






1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users